A Base Station Energy-saving Algorithm Based On Traffic Map In 5G UDN

As the age of mobile internet approaching,mobile communication technologies are evolving to the 5th generation.While mobile data traffic load is growing at an explosive speed,it brings about severe problems of energy consumption.Ultra-dense network is one of the key techniques for 5G to achieve performance expectations like 1000 times growth in system capacity.A large number of small base stations are deployed densely in ultra-dense network to improve spectrum efficiency and system capacity.Meanwhile,the amount of small base stations make the problem of energy consumption even worse,since base stations contribute a large portion in the total energy consumption of communication systems.Thus the study of energy saving techniques in ultra-dense network is of great significance.There are obvious imbalance in the distribution of mobile data traffic loads.If we can make use of this characteristic to flexibly control small base stations with switch between ready state and low power sleep state,energy consumption could be effectively reduced.From this point of view,this thesis proposed a traffic map generating algorithm through historical data of traffic loads,and moreover,a base station sleeping energy saving algorithm.Since the high density and short inter-station distance between base stations in ultra-dense network,the traditional hexagonal cell model is no longer applicable.In this thesis the Voronoi tessellation is used to divide the service cover areas of small base stations.Then the concept of gridding and Sibson interpolation is combined to compose the traffic map generating algorithm,in which the traffic load in every grid could be calculated and the spatial distribution could be illustrated visually.On the basis of traffic map generating algorithm,a base station sleeping energy saving algorithm in ultra-dense network is proposed.First,a resource allocating scheme base on graph theory is used to allocate bandwidth resources for small base stations.Then small base stations are properly controlled to switch between ready state and sleep state on the premise of system traffic requirements being satisfied.By this way the number of sleeping base stations could be as much as possible,leading to lower system energy consumption.This thesis used MATLAB to simulate the traffic map generating algorithm and base station sleeping energy saving algorithm.The results show that the traffic map generating algorithm could illustrate the spatial distribution of traffic loads and the base station sleeping energy saving algorithm could effectively reduce system energy consumption.